AI abuse campaign targets Muslim women in India, researchers say

Muslim women in India are being targeted with AI-made sexual images and videos, according to researchers, online safety workers and women interviewed by Al Jazeera. The abuse matters because fabricated material can damage reputations, deter work and deepen fears of offline harm, even when the images are fake.

Al Jazeera reported that Samreen Ayoub, a 24-year-old freelance model from India-administered Kashmir, discovered last year that an Instagram video had used photos from her student years at Jamia Millia Islamia University in New Delhi. The clip, which included an AI-generated narration and captions styled like a news segment, falsely accused her of selling sex to Hindu men and misidentified her brother as her “pimp”, Ayoub told Al Jazeera.

Ayoub said the video spread through multiple accounts within hours, bringing abusive comments, threats and attacks on her character. She told Al Jazeera that modelling work later dried up because brands were put off by the comments on her social media profiles.

The Washington-based Center for the Study of Organized Hate said it studied 1,326 publicly available AI-generated images and videos posted by 297 public accounts on X, Facebook and Instagram between May 2023 and May 2025. The centre said sexualised AI content depicting Muslim women received more than 6.7 million interactions, the highest engagement in its data set.

Zenith Khan, a co-author of the study and a digital research analyst at the centre, told Al Jazeera that generative tools have made it easy for users with little technical skill to turn hostile ideas into realistic images. The centre said many images in its data paired women coded as Muslim with men coded as Hindu, while portraying Muslim men as violent or immoral and Muslim women as submissive or in need of rescue.

Meri Trustline, an online safety helpline run by the Mumbai-based RATI Foundation, has also reported more cases involving manipulated images. Al Jazeera said the helpline has handled more than 482 cases since it began in 2022, with about 10 percent involving digitally altered material.

Salman Mujawar, a front-line counsellor at the helpline, told Al Jazeera that shame, fear and trauma keep many victims from telling relatives or speaking publicly. Several Muslim women contacted by Al Jazeera declined to speak on the record for similar reasons.

Researchers linked the current AI abuse to earlier attacks on Muslim women online. In 2021 and 2022, the “Sulli Deals” and “Bulli Bai” platforms used doctored images to offer Muslim women in mock online auctions, Al Jazeera reported. Indian authorities arrested Aumkareshwar Thakur, accused of creating the Sulli Deals handle, and Niraj Bishnoi, identified as the creator of Bulli Bai, in January 2022; both later received bail from a New Delhi court on humanitarian grounds.

Atif Rasheed, a politician with the governing Bharatiya Janata Party, told Al Jazeera that AI can be used for good or harm and called for tighter rules against misuse. He rejected viewing the problem as religious, saying the BJP respects women of all faiths and that the Sulli Deals and Bulli Bai cases were addressed under the law.

Legal experts told Al Jazeera that India’s rules have gaps when abuse is wholly AI-generated. Apar Gupta, a lawyer and founding director of the Internet Freedom Foundation, said Section 66E of India’s Information Technology Act applies to capturing or publishing private body images without consent, but may not cover an image that was created artificially and never recorded from the person’s body.

Gupta said platforms can keep “safe harbour” protections if they remove illegal material after notice, but victims often struggle to report deepfakes and prove the content targets them. Ayoub told Al Jazeera she filed complaints with New Delhi’s cybercrime unit and that most abusive posts came down only after friends reported the accounts in large numbers.

This story draws on original reporting from Al Jazeera.